The Three-Body Problem of the Edge: Federated Discovery and Planning for Multi-Operator 5G Networks
Public edge computing requires more than deploying application workloads close to users: the mobile network must also provide a suitable user-plane path toward the selected edge environment. This creates a design-time planning problem involving Mobile Network Operators (MNOs), Edge Computing Service Providers (ECSPs), and Application Service Providers (ASPs). While MNOs control network-side feasibility and ECSPs expose candidate edge environments, the ASP must decide where deployment is viable across regions, operators, and service requirements. Existing standards, including 3GPP NEF, CAPIF, EDGEAPP, ETSI MEC, and GSMA Open Gateway/CAMARA, provide important mechanisms for API exposure, runtime traffic steering, edge service discovery, and endpoint selection. However, they do not provide a standardized way for ASPs or authorized aggregators to evaluate deployment feasibility before infrastructure is placed. This article proposes the Edge Deployment Feasibility Service (EDFS), a standards-aligned northbound interface for exposing abstract, planning-grade feasibility information without revealing sensitive operator topology. EDFS allows candidate deployments to be evaluated using attributes such as target geography, application profile, planning constraints, service class, viability status, limitations, and validity. The article analyzes the resulting standards gap, defines design requirements for EDFS, presents a multi-operator architecture and information model, and discusses possible standardization paths across 3GPP, CAMARA/Open Gateway, and ETSI MEC. EDFS complements existing runtime mechanisms by providing the missing design-time layer between application placement and session-level traffic steering.
Authors
- Spyros Denazis (ORCID: https://orcid.org/0000-0001-6771-3038)
- Dimitris Giannopoulos
Institutions
- University of Patras (GR)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-07-01
- DOI
- https://doi.org/10.5281/zenodo.21107157
- Primary Topic
- Software-Defined Networks and 5G
- Type
- preprint